© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 24016
    Prev
    Next
    INDEX
    Explanations

    information related to organized lists and structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.20
    0.9%
    124
    +0.12
    0.5%
    674
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    124
    +0.20
    0.04
    326
    +0.12
    0.04
    314
    +0.09
    -0.00
    Negative Logits
    theless
    -0.80
     externally
    -0.72
    interrupted
    -0.70
    omething
    -0.69
    enegger
    -0.68
     undermin
    -0.66
     disg
    -0.65
     aven
    -0.65
     latter
    -0.64
    kinson
    -0.64
    POSITIVE LOGITS
     Description
    0.96
    verages
    0.91
     Ratings
    0.89
    alyses
    0.86
    fixes
    0.79
    earances
    0.78
    appings
    0.76
     Rankings
    0.76
     Reviews
    0.75
    autions
    0.74
    Activations Density 0.247%

    No Known Activations